Increased expression of the Fibroblast growth factor 8 (Fgf8) gene apparently plays an important role in the progression of both breast and prostate cancer. To understand how abnormal Fgf8 expression affects cell function, we are studying its normal role during vertebrate embryogenesis, using the mouse as a model system. Fgf8 is expressed in a variety of regions of the embryo that may be termed organizers: regions that are a source of signals that pattern and thus organize the surrounding tissue. An allelic series generated at the Fgf8 locus (Meyers et al. 1998 Nature Genetics 18:136), as well as Cre-mediated tissue-specific knockouts (Lewandoski et al. 2000 Nature Genetics, in press) has revealed a role for Fgf8 in organizers that control gastrulation, limb, and brain development. One of the intriguing insights that has emerged from these studies is that at different stages of embryogenesis FGF8 signaling plays different roles in cell migration, proliferation, patterning, and survival. How is this diversity of response achieved? To answer this question, we are using several approaches to identify and study downstream targets of FGF signaling. Among the approaches we are using are various gene-trap strategies, which provide tagged insertional mutagenesis. The logic that we are applying in adapting these mutagenesis strategies to isolate genes that lie specifically in the FGF signaling pathway can potentially be applied to any genetic pathway. One target gene of FGF signaling is the homeobox gene Gbx2. Gbx2 is expressed in all three germ layers of the late gastrula/early neurula stage of embryogenesis. However, in Fgf8 null homozygotes (which fail to express another Fgf gene, Fgf4), Gbx2 is not expressed, demonstrating that Gbx2 is a downstream target gene of Fgf8, Fgf4, or both (Sun et al. 1999 Genes Dev. 13:1834). Inactivation of the Gbx2 gene has revealed that it is required for normal mid/hindbrain development, playing a role in the positioning of the isthmic organizer, a region that is a source of signals that pattern these regions of the brain (Wasserman et al. 1997 Development 124:2923). We are currently extending this analysis by studying a hypomorphic (partial-loss-of-function) Gbx2 allele. We are also performing Cre-mediated tissue-specific inactivations of Gbx2 to determine exactly where and when this gene is required for normal brain development.